Posit numbers make more efficient use of the bits representing real numbers than do conventional IEEE floating point numbers. This post explains what posits are, how their bits are interpreted, their dynamic range, etc.